JP2010512106A - Announcement Media Processing in Communication Network Environment - Google Patents

Abstract

In order to effectively handle switching between user media and announcement media, the basic step (S1) is first to determine the configuration of the user media. Next, the configuration of the announcement media to be presented is determined based on the determined user media configuration (S2). Subsequently, the announcement media is configured according to the announcement media configuration (S3), and the finally configured announcement media is transmitted to the intended user (S4). In this way, the overall appearance or sound of the announcement is substantially the same or at least similar to the overall appearance or sound of the user media and is preferably undistorted. This allows the user to perceive the announcement as clearly as possible.
[Selection figure] None

Description

  The present invention relates generally to announcement technology in modern communication networks, and more particularly to processing of announcement media and issues related to switching between user media and announcement media. .

  Announcements play an important role in communication services such as telephony services. Announcements are usually pre-recorded media or machine-generated media. The machine-generated media can be generated, for example, with a function for converting text into speech (also called speech synthesis) or a function for converting text into images. Announcements may be generated within the communications network, or may be generated by a remote user switchboard or computer.

  Examples of the use of announcements from communication networks include:

  An error message when the user-initiated command cannot be completed. For example, if the caller has hidden the phone number and the caller has configured not to answer the phone without seeing the phone number, the system will present an error message to the caller. There must be.

  • If the user puts the session on hold, the system can play a hold message or tone on the other party.

  -In a conference call, when a new user joins a session or when a user leaves the session, the conference server may, for example, “John Smith joins the conference” and “John Smith Announcement such as “I was withdrawn from” can be presented.

  -The user has paid upfront and is about to use up the upfront payment. Operators may limit their use due to low balance and want to announce about it at the start of a session or during a session (which may be a very long session).

  An increasingly used method on the Internet is to present an image with a pin code (or password) on a web page. The pin code image is quite distorted, so an automatic text recognition system should not be able to detect the pin code, but a sensible person should still be able to read letters and numbers. This is used in (insecure) email instead of sending the corresponding pin code.

  Examples of caller side announcement usage include:

  A user calls a travel agency to reserve a ticket. The following scenario is assumed.

  1. Users consult with travel agency staff to find the best travel plan. In this step, the discussion takes place between two people.

  2. After making a decision about the trip, the user is asked to key in their credit card number. This is a human-machine communication where the user listens to a pre-recorded message or a machine-generated message and presses the dial buttons (0-9) to enter his / her credit card number. In this process, the following statements are assumed: "Please key in your credit card number", "The number is 1234 5678 9012 3456", "If this is correct, press 1", if not, press 2 "," Your credit Please enter the expiration date of the card ”,“ The expiration date is January 1, 2007 ”. These sentences will be generated by the announcement server.

  3. After keying in the credit card number and other necessary data, the session with the travel agency staff continues to determine further travel options.

  4). These steps may be repeated multiple times.

  • The user calls a service desk or customer contact to solve some problems after purchasing a new computer. The service desk uses a screening process to classify the problem and then call the technician. In this screening process, the user must answer the following questions: “If your problem is software related, please press“ 1 ”and #. If your problem is hardware related, please press“ 2 ”. If you are not sure, please press“ 3 ”.” After this screening process, the session is suspended and a waiting message is played to the user. The technician may answer the call and the waiting message may be interrupted at any time.

  Traditionally, the generation of information messages such as prompts and voice announcements uses conventional pulse code modulation (PCM) or adaptive differential PCM (ADPCM) for encoding and decoding. It is implemented by a relatively simple time division multiplexing (TDM) based announcement machine of a circuit switched system. In modern and future communication systems, the conditions and requirements for processing announcements will change dramatically, so there is a general need to provide a solution for effectively processing announcement media in such communication systems. There is.

  The present invention overcomes the above and other shortcomings of the prior art devices.

  It is a general object of the present invention to improve the processing of announcement media in a communication network environment.

  One object of the present invention is to effectively handle switching between user media and announcement media without at least one of discomfort and distortion associated with transition, or at least reducing distortion caused by switching. It is.

  In particular, it is desirable to allow the user to perceive the announcement as clearly as possible.

  In particular, one particular objective is to improve and provide a method and system for switching between user media and announcement media.

  Another specific object is to improve and provide an announcement server for communication networks.

  The above and other objects are achieved by the present invention as defined by the appended claims.

  Because of the wide variety of media configurations that modern communication networks provide, the overall sound or appearance of the announcement media can be very different compared to the sound or appearance of normal user media, so it is an announcement server. The inventors have recognized that this can cause problems. This can be very uncomfortable for the user.

  In order to handle switching between user media and announcement media, the basic concept of the present invention is to first determine the configuration of the user media, and then based on the determined user media configuration, the announcement media to be presented. It is to determine the configuration. Subsequently, the announcement media is configured according to the announcement media configuration, and the configured announcement media is transmitted to the intended user. In this way, the overall appearance or sound of the announcement is substantially the same or at least similar to the overall appearance or sound of the user media, preferably without distortion and the user perceiving the announcement as clearly as possible Will enable.

  However, it should be understood that the present invention is not limited to voice and audio, but may be applied to video.

  Typically, the user media comes from the remote user with whom you are talking, and the announcement media comes from the announcement server or processor. However, the announcement server may be located on the network side as a network-based announcement server, or it may be located on the user side associated with the remote user, eg, user equipment or a private branch exchange (PBX).

  In a preferred exemplary embodiment of the present invention, to identify one or more valid user media configurations that may be selected or otherwise determined from its user media configuration. The setup of the communication session regarding the user media between the concerned users is monitored.

  Preferably, user communications are monitored to detect the currently used user media configuration, allowing the announcement media configuration to match the current user media configuration. The configured announcement media is then preferably transferred to the intended user by causing the announcement server to insert the configured announcement media into the session. Optionally, the announcement media is transmitted in parallel with the user media, for example by establishing a new session for the announcement media.

  Preferably, the codec and / or transmission format configuration for the user media is determined, and then the announcement media configuration includes the announcement media codec and / or transmission format configuration and the user media codec and / or transmission format configuration, and It is determined based on the matching (matching).

  In another aspect, the invention relates to an announcement server for a communication network. Basically, the announcement server configures the announcement media to be inserted into the session based on the media configuration information so as to obtain the media configuration information indicating the configuration of the user media in the network communication session. Configured as follows. In addition, the announcement server is operable to insert the configured announcement media into the session.

  Other advantages provided by the present invention will be appreciated upon reading the following description of embodiments of the invention.

  The invention, together with other objects and advantages thereof, will be best understood by reference to the following description taken in conjunction with the accompanying drawings.

FIG. 1 is a schematic diagram illustrating an example system overview. FIG. 6 is a schematic diagram showing distortion when the encoder state is reset while the decoder state is not reset. FIG. 6 is a schematic diagram showing distortion when the encoder state is not reset while the decoder state is reset. FIG. 4 is a schematic flow diagram of a basic method of switching between user media and announcement media according to an exemplary embodiment of the present invention. FIG. 6 is a schematic flow diagram of a method for switching between user media and announcement media according to another exemplary embodiment of the present invention. FIG. 2 is a system overview including a network side and a user side according to an exemplary embodiment of the present invention. FIG. 4 is an overview according to another exemplary embodiment of the present invention. FIG. 3 is an overview diagram according to yet another exemplary embodiment of the present invention. FIG. 2 is a schematic block diagram of an announcement server according to a possible preferred embodiment.

  Throughout the drawings, the same reference characters are used for corresponding or similar elements.

  For a better understanding of the present invention, it may be helpful to start with a brief system overview.

  FIG. 1 is a schematic diagram showing an overview of an exemplary system. This exemplary system includes two end user A and B user equipment (UE) 10-1, 10-2 communicating via associated base stations 20-1, 20-2 and a network-based announcement server. 30. In this case, a media stream, such as a real-time protocol (RTP) [1] media stream, always passes through an announcement server which may be a media resource function processor (MRFP) [2]. In this example, the announcement server replaces the RTP stream from user A with an announcement, which is called an in-band announcement.

  Other ways of introducing announcements are possible, for example, where media is sent directly between A and B without going through the announcement server. In this case, the announcement may be sent in SIP INVITE from the announcement server to the UE (User Equipment) B, and the UE B receives the media received from UE A to generate the media received from the announcement server. Must be destroyed. Another alternative is to send a message such as a SIP INFO or SIP NOTIFY message that includes a link (URL) to the announcement media. In general, an alternative method is thus to send announcement media in parallel to the user media, with or without establishing a new session for the announcement media.

  The announcement server may be installed in the communication network. In modern systems such as the IP Multimedia Subsystem (IMS), the announcement server will typically be installed in the Media Resource Capability Processor (MRFP), but could also be installed in an application server such as a conference server, for example. .

  It is also possible to have an announcement server in or near the endpoint, in user equipment, in a computer, or in a private branch exchange (PBX).

  Careful analysis by the inventors has revealed that existing solutions are exposed to one or more of the problems described below.

  At present, in existing circuit-switched systems, the announcement works, which works well because the codec used is typically PCM [4] or ADPCM [5]. These codecs are sample-by-sample codecs that do not use prediction at all (PCM) or use only a very limited amount of prediction (ADPCM). This means that if the decoder recovers very quickly from a state mismatch, it is unlikely to cause audible distortion.

  Furthermore, conventional systems use only one codec, for example using either PCM or ADPCM, and never both. The system also uses the same transmission format during all sessions, for example, never changes codec rate, frame aggregation, or redundancy depending on the situation. In fact, the system uses the same configuration for all sessions.

  In modern and future systems such as the IP Multimedia Subsystem (IMS), the situation is completely different, especially with regard to Multimedia Telephony (MMTel) [6]. Several types of media can be sent. This media may be encoded with different codecs and at different rates. Different transmission formats may be used. That is, the payload format may be used with or without frame aggregation and with or without redundancy. These variations are necessary to reduce the risk to network congestion and to adapt the session so that it can be maintained with reasonable quality even during periods of congestion. This is particularly important in cellular systems where the radio channel changes significantly. However, these variants cause problems for the announcement server. It is not possible to encode the announcement media in just one way as it did before and hope that the recipient is satisfied. Media between UE A and UE B is typically encoded with a wideband codec (AMR-WB), but if the announcement is encoded with a narrowband codec (adaptive multirate: AMR), the media is completely It will sound different, and the receiving user will be wondering why the announcement sounds so different from normal media. In the worst case, the receiving user may become uncomfortable and more interested in media quality than the actual content of the announcement, which means that the user may even misunderstand the announcement.

  In addition, the use of modern prediction-based codecs can lead to state discrepancies when announcements interrupt regular media, and can also cause audible distortions that can be uncomfortable for the user. There is. Prediction is very important for modern codecs such as AMR [7] or AMR-WB [8]. Inter-frame prediction is used to reduce the bit rate, ie to achieve a high compression ratio, but still provides good quality. In inter-frame prediction, the state needs to be passed from frame to frame. If the announcement interrupts normal media, two different codec instances are used, one codec instance for UE A for voice media from the user and one codec instance for the announcement server, resulting in a state mismatch. I will. The state of UE A is evolving according to the prediction used, while the state of the announcement server starts from the default state. State mismatch can cause more or less audible distortion, depending on the current content. Two examples of such distortion are shown in FIGS. In both cases, the distortion is clearly audible and easily noticed by the listener, but the spike in FIG. 2 is much more disturbing.

  2 and 3, it can be understood that it takes about 100 to 200 ms for the synthesized sound to recover to a normal state after the asynchronous reset. Instead, a stateless codec such as PCM will soon return to a normal state as it does not have to “make up” the state with the appropriate content.

  Conventional circuit switching systems typically also have control of the volume level, and adjust the volume within the network if the volume is not appropriate. VoIP systems such as multimedia telephony will probably not have such functionality, since the idea is to send VoIP packets end-to-end without transcoding and other types of changes. . With respect to VoIP, therefore, if the other party is speaking loudly or quietly, the end user is likely to adjust the volume considerably. If the announcement server inserts the announcement media without checking the volume, it will be presented at a volume that is too low, or it will be presented at a very high volume and the listener will have to take the phone off the ear immediately, so the entire message It may be overlooked.

  These problems are not limited to voice. Similar problems occur with audio and video. With respect to these issues, since the codecs for these media types typically have a higher compression ratio than the voice codec, and depend even more on good quality conditions to achieve this compression ratio, Greater problems can be anticipated.

  Embodiments of the invention relate to one or more of these issues.

  First, an example embodiment will be described that is primarily directed to the problems posed by the wide variety of possible media configurations of modern communication networks. The problem is that the overall sound or appearance of the announcement media can be very different compared to the sound or appearance of normal user media. This can be very uncomfortable for the user.

  FIG. 4 is a schematic flow diagram of a basic method of switching between user media and announcement media according to an exemplary embodiment of the present invention. In order to effectively handle switching between user media and announcement media, the basic step (S1) first determines the configuration of the user media. Next, the configuration of the presented announcement media is determined based on the determined user media configuration (S2). Subsequently, the announcement media is configured according to the announcement media configuration (S3), and the configured announcement media is finally transmitted to the intended user (S4). In this way, the overall appearance or sound of the announcement will preferably be substantially the same or at least similar to the overall appearance or sound of the user media without distortion. This allows the user to perceive the announcement as clearly as possible. For example, user media and announcement media may be voice, audio or video.

  Preferably, in the exemplary embodiment of the present invention, the codec configuration of the user media is determined, and then the codec configuration of the announcement media is matched with the determined codec configuration of the user media. This means that the same or at least a similar codec configuration is used for the announcement media. This increases the chance that the two media will sound or look similar to the intended user. For example, the codec configuration may also include a codec type, a codec mode, and optionally a codec mode switching function.

  It is also possible to determine the transmission format configuration of the user media and to match the transmission format configuration of the announcement media with the transmission format configuration of the user media. In this way, the announcement media is not likely to be severely affected by transmission failure.

  Optionally, the user media frame aggregation and redundancy format configuration is determined, and the announcement media frame aggregation and redundancy format configuration may be matched with the user media announcement media frame aggregation and redundancy format configuration. It becomes possible.

  FIG. 5 is a schematic flow diagram of a method for switching between user media and announcement media according to another exemplary embodiment of the present invention.

  Preferably, the setup of a communication session for user media between the user devices of the relevant user is monitored to identify one or more valid user media configurations from which an appropriate user media configuration can be selected or determined. (S11). Subsequently, user communication during the session is monitored to detect the currently used media configuration in the valid user media configuration (S12). Next, the announcement media configuration is matched with the current user media configuration (S13). Optionally, the appropriate timing for inserting the announcement media during the session is determined, for example, by considering the urgency of the announcement (S14). By causing the announcement server to insert the configured announcement media into the session (S15), the configured announcement media is finally transferred to the intended user. Optionally, the announcement media is sent in parallel with the user media, for example, by establishing a new session for the announcement media or simply sending a control message with a link to the announcement media. .

  In this particular scenario, the basic definition of the session is normally negotiated during session setup, for example by using Session Initiation Protocol (SIP) Session Description Protocol (SDP). For example, several possible valid media configurations may be defined with SDP signaling and associated with each identifier. In certain exemplary embodiments, the payload type field of the RTP packet may be used to identify the media configuration by utilizing the payload type number as an identifier. This number can be associated with a valid media configuration. The association between the payload type number and the media configuration is preferably performed at least one time during session setup (eg, SIP INVITE) and during session reconfiguration (SIP UPDATE or so-called SIP RE-INVITE). During subsequent communication, upon transmission of a media packet, such as an RTP packet, the payload type field may be monitored to extract the payload type number, and then the payload type number is determined based on the currently used media configuration. Can be associated with

  As an example, a media configuration may be defined by one or several of the following information items:

Codec (group)
For example, AMR, GSM-EFR (GSM Enhanced Full Rate), EVRC (Enhanced Variable Rate Codec), etc.

Codec mode (group)
Where applicable For example, for AMR, all eight codec modes may be allowed, or a subset of codec modes may be allowed.

Codec mode switching function When applicable For example, it may be specified for AMR that codec mode switching is allowed only between adjacent codec modes, ie, bit rates 12.2, 7. Subsets of AMR codec modes such as codec modes specified at 4, 5.9 and 4.75 kbps are allowed, and switching from a code rate mode of 12.2 to a codec mode of bit rate 4.75 is desired. If so, the switch must go through codec modes 7.4 and 5.9.

Payload format Where applicable For example, there are two basic options for AMR: bandwidth efficient and octet-aligned.

• Recommended number of frames per packet This is generally not a strict requirement.

• Maximum amount of data per packet This is generally a strict requirement.

  Therefore, during session setup, various possible media configuration ranges are usually specified. By just selecting any media configuration among the media configurations negotiated during session setup, it is possible to obtain a reasonable but usually sub-optimal solution. If several valid media configurations are allowed, a better solution may be to provide a priority order that indicates in which order the various configurations should preferably be used. For example, if exemplary configurations A, B, and C are allowed, the preferred order in SIP INVITE can be specified as B, C, A.

  However, by monitoring which configuration (s) are currently being used during user communication, it is possible to optimize the selection of media configuration for the announcement media.

  As an example, robustness can be achieved by the user client perceiving a bad channel, eg, using the lowest possible bit rate and adding the same frame several times to add redundancy. Is best applied when the announcement is processed using the same or similar configuration. Otherwise, the default configuration for speech coding (encoding) will usually start at the highest possible bit rate and then perform adaptation in the direction that will be more robust. However, this adaptation is usually time consuming and there is a risk that the announcement is already completed before the adaptation reaches the same level of robustness corresponding to that used for the user media. Using the present invention, the announcement media configuration can be immediately matched to the user media configuration by monitoring the currently used media configuration.

  If the announcement server does not support some or all of the valid media configurations shown during session setup for any reason, a local decision on which media configuration to use is allowed by the announcement server Based on (a subset of) the configuration, it may be performed on the announcement server.

  In general, user media and announcement media are directed to a first network element, such as a first user's user equipment. User media typically comes from a second network element, such as a remote user's user equipment. Announcement media usually comes from a third network element such as an announcement server. However, it should be noted that the announcement server may be located on the network side as a network-based announcement server or may be located on the user side associated with the remote user. In the latter case, the second network element and the third network element may be the same or at least close to each other. Various exemplary embodiments of the invention are described.

  FIG. 6 is a system overview including a network side and a user side according to an exemplary embodiment of the present invention. Basically, user equipment (UE) 10-1, 10-2 of two users A and B are communicating via a communication network. In this particular example, the communication passes through the announcement server 30, which inserts a properly configured announcement into the communication between users. The announcement server 30 includes a module 32 for determining user media configuration and announcement media configuration, a database 34 holding one or more pre-recorded announcements or machine-generated announcements, and a module 36 for configuration of announcement media. And a controllable switching mechanism 38 for replacing the user media stream with the configured announcement media.

  Preferably, the announcement server 30 is configured to monitor session setup between users and identify one or more valid user media configurations. Information about valid user media configurations is typically stored with associated identifiers in a table (not shown) associated with the announcement server. The announcement server 30 may then select from these media configurations in a number of different ways. In a preferred exemplary embodiment of the present invention, user media packets are monitored during a session to identify the currently used user media configuration. Preferably, this extracts a media configuration identifier from the packet header of one or more media packets and maps this identifier to a specific user media configuration stored in a valid user media configuration table. Executed by.

  Based on information about the specified user media configuration, it is then possible to determine the appropriate configuration of the announcement media. Once the announcement media configuration is determined, the selected announcement obtained from the database 34 or generated by other means may be configured in the configuration module 36 accordingly. This typically includes at least one of encoding and formatting of the announcement media according to the determined configuration. The announcement media may then be inserted into the communication session. Although details will be described later, it may or may not be necessary to determine an appropriate timing for inserting the announcement depending on the situation.

  It may also be advantageous to continuously monitor the control information during a communication session between users. In particular, it may be advantageous to monitor feedback information for link adaptation to identify any possible changes to the user media configuration, such as changes in codec mode or redundancy mode, which Enables adaptation of announcement media configuration according to the latest information.

  The monitoring may be executed by the announcement server 30 or the option unit 40 related to the announcement server.

  FIG. 7 is an overview diagram according to another exemplary embodiment of the present invention. In this example, the user media is transmitted between the users 10-1 and 10-2 through the network, but does not pass directly through the announcement server 30. Preferably, the external monitoring unit 40 also monitors at least one of session setup and optionally user communication and control feedback to cause the determination module 32 to determine the user media configuration and the announcement media configuration. Then, according to the determined announcement media configuration, the configuration module 36 configures an announcement that is obtained from the announcement database 34 or generated by other means. The configured announcement may then be sent from the output module 38 to the intended user, for example by using SIP INVITE, etc., to establish a new session for the announcement media. The intended user, eg, User B, must then typically discard the user media from User A to generate the announcement media unless multiple sessions are supported by the user equipment.

  FIG. 8 is a schematic overview according to yet another exemplary embodiment of the present invention. In this example, the announcement server exists on the user side. In addition to the usual equipment for inter-user communication, the entire user equipment (UE) system 10-1 of user A also comprises an announcement server (AS) or equivalent announcement device that can provide announcement media. The entire user equipment system further includes a media configuration unit (MC). Based on the information about the configuration of the user media, the MC unit then determines an appropriate announcement media configuration and configures the announcement to be presented to the other party user B 10-2 accordingly. In the entire user system, the user media may be switched to the announcement media and then switched back.

  In another alternative embodiment, the announcement server is implemented in a private branch exchange (PBX) connected to the endpoint.

  In fact, there may be more than one announcement server in the network system, and the present invention also switches between announcement media from the first announcement server and announcement media from the second announcement server. It can be processed in the same or similar manner as described above for switching between user media and announcement media. For example, user media from the UE may be replaced with announcement media from the PBX announcement server, and announcement media from the PBX announcement server may be replaced with announcement media from the network-based announcement server.

  In the following, further exemplary embodiments will be described on how switching between media from the caller (human) user and media from the announcement server can be managed.

  For example, a user may be involved in multimedia telephony that uses IP, UDP and RTP for media transmission and SIP for session control. The announcement server may be located in the telecommunications network or at the remote caller's location.

  The overall goal is to ensure that switching between two media sources and media types does not bother. Preferably, this means that the announcement media should be encoded with a similar codec, similar to the media between UE A and UE B, and transmitted using a similar transmission format. It means that it should be. This also means that the switching should create no distortion or at least as little distortion as possible. This may be done in the following exemplary manner.

Announcement server
1. Monitor session setup and determine which codecs can be used in the session. This is usually static information during the session, but may be changed if any session parameters are renegotiated during the session.

  2. Monitor media packets and / or quality feedback to determine currently used codecs and transmission formats. This is temporary information that usually needs to be updated fairly frequently, and perhaps not every single packet, but at least periodically. The frequency of these updates may depend on the system load. For at least one of high system load and channel condition changes, it may be expected that the media between users will need to be adapted fairly often. However, adaptation should be fairly rare for low system loads.

  3. Determine the appropriate encoding and transmission format for the announcement media.

  4). Determine how to generate appropriate transitions.

    a. Whether to use fade-in and fade-out for announcements.

    b. Whether to add idle periods (silence, blank images).

    c. Whether the decoder needs to be reset (codec homing, ensure CRC is not functioning, etc.)

  5. Determine the appropriate timing to insert the announcement. This timing usually depends on the urgency of the announcement.

  6). Insert announcement and discard media from UE A. The volume may need to be adjusted to match the volume of the media from UE A.

  7). It is also determined how to generate an appropriate transition from UE A to the media after the announcement is complete.

  Note that not all of these steps need to be performed in this exact order. In some cases, it may be possible to skip some of these steps.

  For a better understanding of the present invention, an announcement server according to a possible preferred embodiment will now be described in more detail with reference to FIG.

  In this particular example, the functions of the various blocks can be summarized as follows:

-Session setup between UE A and UE B is preferably done in SIP. This determines which media can be used in the session, how to encode each media, and which transmission format to use. Some examples to clarify are
Whether the media is audio and / or video and / or text;

    -Which codec should be used for each media? Several codecs may be allowed for one medium. Session setup may conclude, for example, that both AMR and AMR-WB are acceptable for voice in the session. A codec may also include several codec modes. For example, AMR defines eight codec modes. Each codec mode has its own bit rate (12.2, 10.2, 7.95, 7.4, 6.7, 5.9, 5.15 and 4.75 kbps). Some video codecs define “profiles” and “levels” that determine how video should be encoded instead.

    -Which payload format should be used? Each codec has its own payload format. Some payload formats define several variations of the payload format. For example, for AMR and AMR-WB, the payload format defines both a bandwidth efficient format and an octet array format.

    -How can the codec mode switch be performed if the encoder is free to switch to any negotiated mode at any time? For example, for scenarios of interaction with circuit switched services for AMR in UTRAN and GERAN, such codec mode switching is limited to meet the constraints that exist in at least one of the UTRAN and GERAN CS networks. Sometimes it is necessary.

    Thus, media may typically be configured in several different ways. One RTP payload type number is defined for each configuration allowed for the session. Note that you may expect several configurations for each media.

  Media input is media from UE A (voice, audio, video, text, etc.). Optionally, the media could be media from another announcement server as described above. This media will be interrupted by the announcement. RTP is usually used for transmitting media.

  Media output is media sent to UE B.

  UE B sends a quality feedback or adaptation request to UE A. Quality feedback is typically sent over RTCP and is often in the form of numerical indicators (packet loss rate, jitter, etc.). Examples of adaptation requests include codec mode requests, frame aggregation requests, redundancy requests, and intra refresh of video frames. The feedback is used by the sender (UE A) to adapt the bit rate, packet rate and redundancy to reduce the media impact of transmission dysfunction.

    Note that you do not know whether there is a transmission constraint that triggers adaptation between UE A and the announcement server, or whether it is between the announcement server and UE B. However, it is beneficial for the announcement to be encoded and formatted in the same way as the media from UE A in order to minimize the risk that the announcement will be severely compromised by transmission failure between the server and UE B. is there.

  The session and media analysis unit analyzes session setup and / or media and / or media feedback to detect the characteristics of the input media and the currently used transmission format. Examples of detected characteristics include the following.

-Media type (group)
-Codec (group), codec rate (group) or codec mode (group), frame aggregation, redundancy, etc.-Payload format (group)
-The sound level of the media, ie the volume. This probably requires media decoding. In order to avoid running the decoder without a break, this analysis should be done less frequently and preferably only once or a few times during the session.

    Whether the media is idle, eg silence, background noise, blank images, etc. One way to detect this is to verify whether a SID UPDATE frame is currently being transmitted, which requires the RTP packet to be parsed (or at least the size needs to be checked). However, it does not require media decoding.

Whether the media is in an active state, e.g. active voice, video, etc. In that case,
Whether the media has been constructed or just started, eg immediately after the start of audio.

      Whether the media will fade out, this happens at the end of sentences and words.

    -Note that depending on the situation, not all parts of this analysis need be performed continuously or at all. Some apply only when session setup and session are renegotiated, while others are preferably monitored during sessions (with varying degrees of frequency). Part of this analysis may require decoding of RTP packets.

  The media database contains actual announcement media (voice, audio, video, images, etc.).

Media encoders and modifiers have the ability to encode and format at least one of the media as well as media from UE A, eg
-Various codecs (AMR, AMR-WB, H.263 [9], H.264 [10], etc.) may be used.

    -The acoustic bandwidth may be changed (filter to narrowband, filter to wideband, bandwidth extension, screen size, fade-in and fade-out application, etc.).

    Whether an idle period (silence, blank image) should be added at the start and / or end time).

    -Codec homing information may also be added at the start and end of the announcement to reset the decoder.

    -Adjust the volume level.

    -Use various payload formats.

    Apply frame aggregation and redundancy when used by UE A.

  -The announcement command notifies the following information to the announcement server.

    -Which announcement should be used?

    -The urgency of the announcement, i.e. how important it is to present the announcement immediately to the user, or whether the announcement may be delayed for a while.

  The list of possible media change methods is a description of which media change method is appropriate for each state as determined by the media analyzer. Examples of various operations appropriate for various states are described below.

  The controller receives the announcement command and information about how the media is currently formatted in the session between UE A and UE B.

    -If the controller should change it in any way before sending the media, how the media should be encoded and how the encoded media should be formatted into RTP packets judge. This information is sent to the media encoder and modifier.

    The controller also determines when the media from UE A is interrupted and when announcement media is inserted.

  If the urgency is high, the media from UE A should be interrupted immediately or with the shortest possible delay.

  If it is less urgent, it should wait until it detects an idle period in the media from UE A.

  An announcement command also triggers an announcement. This trigger may occur from several different locations, as illustrated in the background section. As an additional example, a telephone operator who may want to interrupt the call to notify one of the callers that an emergency call is waiting, or wants to make a "smooth interruption" Includes network operators who may do so. For example, if one of the callers moves out of the service area, the operator may desire to insert an announcement of the type “the call partner has moved out of the service area” instead of simply interrupting the service.

Exemplary Media Encoding and Modification Methods Encoding and Transmission Format The controller should match the encoding and transmission format to the current media encoding and transmission format being transmitted between UE A and UE B.

  The controller should select the same codec and codec mode / rate as the codec and codec mode / rate currently used in the session.

    If the announcement media has a wider acoustic bandwidth, it should be passed through a band pass filter to match the selected codec.

    -If the announcement media has a narrower acoustic bandwidth, bandwidth extension should be applied.

    -If the exact same codec or codec mode / rate cannot be selected, codecs with similar characteristics (acoustic bandwidth, bit rate, frame rate, encoding quality, etc.) should be selected.

  The controller should select the same frame aggregation and redundancy format as the frame aggregation currently used in the session (number of frames per packet) and redundancy format (repetition of frames every few RTP packets).

  The controller should select the same transmission format as the transmission format currently used in the session (RTP payload type, payload format, payload format version).

  Choosing an appropriate encoding format increases the chance that the two media will sound virtually identical.

  Selecting an appropriate transmission format increases the chance that the announcement media will not be severely affected by transmission failure.

Smooth transitions when switching If the announcement message is not very important and if the announcement server keeps the message until it detects an idle period in the voice, it is special if the media switches from different sources. Smoothing should not be required.

  If the announcement message is very important, the announcement server will normally suspend even if the media from UE A is active. In this case, the announcement server should:

  Apply fade out to media from UE A (this may not always be possible).

  Format the announcement media so that it resets the decoder's state or operation of the ECU for several frames, for example, by triggering a reset of the decoder in stages.

    For AMR, this can be done by inserting a codec homing frame before the announcement media starts. Other codecs may not have similar characteristics.

    Ensure that CRC or checksum verification fails. Not all payload formats contain such information. This solution needs to be repeated for several frames to ensure that the signal is sufficiently weakened.

    -Set the bad frame indicator (BFI) bit, which triggers the operation of the ECU at the receiver. Not all payload formats contain such information. This solution needs to be repeated for several frames to ensure that the signal is sufficiently weakened.

    Insert a dummy frame or idle frame (NO_DATA, silence, blank image). NO_DATA is not defined for all codecs. An announcement server must encode silence frames to achieve this. This solution needs to be repeated for several frames to ensure that the signal is sufficiently weakened.

    -The presence of these methods depends on the codec, so the controller needs to decide which operation applies to the currently used codec.

  -Adjust the size of the announcement media to be about the same as the size of media from UE A.

  -Apply fade-in to announcement media.

  A similar operation should be performed when the transmission of the announcement media is complete and when switching back to the media from UE A.

  Note that smoothing may be required at the start or end of the announcement media, both, or may not be required at all.

  As already explained, the present invention is not limited to announcements inserted in telecommunications networks. A similar announcement exists at most commercial service desks ("Press 1 to continue ordering tickets. Press 2 to change date and time. Press 3 to talk to sales staff"). In this case, the announcement comes from the remote “user” and the remote “user” may switch between the pre-recorded message and the human speaker.

  Distortion due to switching between media will be completely eliminated or at least reduced. The format of the announcement media is preferably matched to the media format used in the conversation. This provides a more comfortable transition between the media from UE A and the announcement media.

  There is also an advantage in complexity in terms of both MIPS and memory at the receiving UE, since it is not necessary to have several active codec instances running in parallel.

  It should also be understood that the present invention is not limited to RTP media streams, is not limited to the use of UDP as a transport protocol, and is not limited to the use of SIP for session setup.

  It should be understood that the embodiments described above are given by way of example only and the present invention is not limited thereto. Further modifications, changes and improvements that maintain the basic principles disclosed herein and defined in the claims are within the scope of the invention.

Abbreviation ADPCM Adaptive Differential PCM
AMR Adaptive Multi-Rate
AMR-WB AMR Wideband (AMR-WideBand)
BFI Bad Frame Indicator
CRC Cyclic Redundancy Code
CS circuit switching (Circuit Switched)
ECU error concealment unit
EVRC Enhanced Variable Rate Codec
GERAN GSM / EDGE Radio Access Network
GSM Global System for Mobile communications
GSM-EFR GSM Enhanced Full Rate
IMS IP Multimedia Subsystem
IP Internet Protocol
MIPS Million Instructions Per Second
MMTel Multimedia Telephony
MRFP Media Resource Function Processor
PCM Pulse Code Modulation
Private branch exchange (PBX)
PS Packet Switched (Packet Switched)
RTCP Real-Time Control Protocol
RTP Real-Time Protocol
SDP Session Description Protocol
SIP Session Initiation Protocol
TDM Time Division Multiplexing
UDP User Datagram Protocol
UMTS Universal Mobile Telecommunication System
UTRAN UMTS Terrestrial Radio Access Network
UE User Equipment
VoIP Voice over IP
References [1] RFC 3550 “RTP: A Transport Protocol for Real-Time Applications”, by H. Schulzrinne, S. Casner, R. Frederick, V. Jacobson [2] 3GPP TS 23.228 “IP Multimedia Subsystem (IMS), Stage 2”
[3] RFC 3261 “SIP: Session Initiation Protocol”, RFC 3261, June 2002, Rosenberg, J, Schulzrinne, H., Camarillo, G., Johnston, A., Peterson, J By Sparks, R., Handley, M., E. Schooler [4] ITU-T Recommendation G. 711 "Pulse Code Modulation (PCM) of Voice Frequencies"
[5] ITU-T Recommendation G. 726 “40, 32, 24, 16 kbit / s Adaptive Differential Pulse Code Modulation (ADPCM)”
[6] 3GPP TS 26.114 "IP Multimedia Subsystem (IMS): Multimedia Telephony: Media Handling and Interaction"
[7] 3GPP TS 26.071 “Mandatory Speech Codec speech processing functions; AMR Speech CODEC; General description”
[8] 3GPP TS 26.171 “Speech codec speech processing function: Adaptive multi-rate wideband (AMR-WB) speech codec: General description (Speech codec speech processing functions; Adaptive Multi-Rate-Wideband (AMR-WB) speech codec ; General description) "
[9] ITU-T recommendation H.264. 263, Video coding for low bit rate communication
[10] ITU-T recommendation H.264, Advanced video coding for generic audiovisual services

Claims (25)

A method for switching between user media and announcement media directed to a first network element in a communication network comprising:
The user media comes from a second network element;
The announcement media comes from a third network element;
The method
Determining a configuration of the user media;
Determining the configuration of the announcement media presented to the first network element based on the determined configuration of the user media;
Configuring the announcement media according to the configuration of the announcement media;
Transmitting the configured announcement media to the first network element. Determining the configuration of the user media includes determining a codec configuration of the user media;
The method according to claim 1, wherein the step of determining the configuration of the announcement media is based on matching between the codec configuration of the announcement media and the codec configuration of the user media. Determining the configuration of the user media further includes determining a transmission format configuration of the user media;
The method according to claim 2, wherein the step of determining the configuration of the announcement media is further based on a match between the transmission format configuration of the announcement media and the transmission format configuration of the user media. . Determining the user media configuration further comprises determining a frame aggregation configuration and a redundancy format configuration of the user media;
The step of determining the configuration of the announcement medium is further based on matching between the frame aggregation configuration and the redundancy format configuration of the announcement media, and the frame aggregation configuration and the redundancy format configuration of the user media. The method according to claim 2 or 3, characterized in that: The method according to claim 2, wherein the step of determining the codec configuration of the user media includes determining a codec mode switching function according to a codec type, a codec mode, and an option. A communication session is set up for user media between the first network element and the second network element;
The method of claim 1, wherein transmitting the configured announcement media to the first network element includes inserting the configured announcement media into the communication session. Method. The step of determining the configuration of the user media includes:
Monitoring the setup of a communication session for user media between the first network element and the second network element to identify at least one valid user media configuration;
The method of claim 1, comprising determining a configuration of the user media based on a configuration of the at least one valid user media. Multiple valid user media configurations are identified when monitoring the session setup;
8. The method of claim 7, wherein the user media configuration is determined from the plurality of valid user media configurations. The step of determining a configuration of the user media further monitors communication between network elements in the session to detect a currently used one of the plurality of valid user media configurations. 9. The method of claim 8, comprising steps. Monitoring the communication between the network elements includes detecting information for adapting the configuration of the user media in order to achieve an appropriate determination of the configuration of the announcement media. The method according to claim 9. The timing for inserting the announcement media into the session is determined based on the urgency of the announcement,
The method of claim 6, wherein the configured announcement media is inserted into the session according to the determined timing. The method of claim 6, further comprising applying a transition process for a smooth transition between the user media and the announcement media when the announcement media is inserted into the session. . The first network element includes a user equipment of a first user;
The second network element includes a user equipment of a second user;
The method of claim 1, wherein the third network element includes a user equipment of a third user. 14. The method of claim 13, wherein the announcement server is located on one of the network side and the user side. 14. The method of claim 13, wherein the user is involved in a multimedia session that uses a real time protocol (RTP) for media transmission and a session initiation protocol (SIP) for session control. A system for switching between user media and announcement media in a communication session in a communication network,
Means for determining a configuration of the user media for the communication session;
Means for determining a configuration of the announcement media to be inserted into the session based on the determined user media configuration;
A system comprising: means for configuring the announcement medium in accordance with the configuration of the announcement medium; and means for inserting the configured announcement medium into the session. The means for determining the configuration of the user media includes means for determining a codec configuration of the user media in the session,
The system according to claim 16, wherein the means for determining the configuration of the announcement media is configured to match the codec configuration of the announcement media with the codec configuration of the user media. . The means for determining the configuration of the user media further includes means for determining a transmission format configuration of the user media in the session,
The means for determining the configuration of the announcement media is further configured to match the transmission format configuration of the announcement media with the transmission format configuration of the user media. The system described in. The system according to claim 17, wherein the means for determining a codec configuration of the user media in the session includes a means for determining a codec type and a codec mode. The means for determining the configuration of the user media is:
Means for monitoring the session setup to identify at least one valid user media configuration;
The system according to claim 16, further comprising: means for determining a configuration of the user media based on a configuration of the at least one valid user media. The means for determining the configuration of the user media is:
Means for monitoring said session setup to identify a plurality of valid user media configurations;
16. A means for monitoring communications between users in the session to detect currently used ones of the plurality of valid user media configurations. System. Means for determining when to insert the announcement media in the session based on the urgency of the announcement;
The system of claim 16, wherein the means for inserting the configured announcement media is operable to insert the configured announcement media into the session according to the determined timing. . The system of claim 16, further comprising means for applying a transition process for smooth transition between the user media and the announcement media when the announcement media is inserted into the session. The communication session relates to communication between user equipment of different users;
The system of claim 16, wherein the announcement comes from an announcement server located on one of the network side and the user side. An announcement server for a communication network,
Means for acquiring media configuration information indicating a configuration of user media in a session of the communication network;
Means for configuring announcement media to be inserted into the session based on the media configuration information;
Means for inserting the configured announcement media into the session.

Images